Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Simple mechanisms to measure joules output?

  1. Mar 7, 2010 #1
    I'm tinkering with a simple heat engine and I want to experimentally measure its work output.

    The engine pushes a piston once per test, and I know the work done is on the order of 1 kilojoule. The problem is that the engine does the work on the order of milliseconds, but depending on the configuration that duration of work could vary by an order of magnitude.

    Therefore in order to measure total work per test I can't just use a spring to look at the peak force exerted because the duration is not only very short but also can vary significantly. Also the work done per test might vary by a factor of 2 or 3 depending on other variables, so I can't assume work done is constant between tests.

    I hope to find a way to measure work done per test with an error of less than 5%.

    What is the simplest mechanism -- gauge, or otherwise -- that I could construct or employ to measure joules output by this engine?
  2. jcsd
  3. Mar 7, 2010 #2
    I would use a pressure transducer to monitor the pressure inside of the combustion/gas/working fluid temperature. You can find pressure transducers that have a very high response time and accuracy. Since the geometry of your engine is fixed (I'm guessing) the pressure inside the engine with respect to time will allow you to calculate the total work done per cycle.
  4. Mar 7, 2010 #3
    Searching for pressure transducers is only turning up industrial probes that themselves cost hundreds of dollars, and that's before finding or building an electronic system to record the data at the sub-millisecond frequency I would need in order to calculate the integral of pressure over time to get an accurate work number!

    I was hoping for something more suited to the budget and technical abilities of a hobbiest.
  5. Mar 7, 2010 #4
    Force times time is a momentum impulse. Force times distance is work. Pressure times area = force. 1 kilojoule is a Newton-meter (e., g., 10,000-Newton force times a 0.1-meter stroke).

    Bob S.
  6. Mar 8, 2010 #5
    You could use a mechanical indicator to construct a PV diagram, then use a planimeter to measure the enclosed area. Commonly done with steam engines.
  7. Mar 8, 2010 #6
    Ah ha ... I guess what I really want to measure is momentum impulse, not work. The engine does not necessarily create motion on the piston end if the piston is stuck against a large mass. So this is actually more like a rocket.

    So how can I accurately measure impulse if the force is on the order of 1000 Newtons and the duration is on the order of 1ms?

    One possibility I've found is a "ballistic pendulum."
  8. Mar 12, 2010 #7
    Can you explain your system in more detail? For now I think you can use a coil spring which has some marker like a pen on it, so you can measure the maximum compression distance and calculate the variation in force within the spring. At the same time, you can use a microphone to pick up the sound, and try to pull out the duration of the compression on the computer using some music software. Probably from those, you can calculate displacement, force and compression time to calculate impulse,work done,etc.
  9. Mar 12, 2010 #8
    Yes, it's essentially a piston fired by combustion of various volatile fuels. Right now it's configured with an essentially direct-vent exhaust.

    A coil spring will suffice to measure peak force since it's easy to mark peak compression, but firing the engine occurs in milliseconds so I would need microsecond resolution on video to get an accurate read of compression over time.

    What sound input you were suggesting I analyze to measure instantaneous force? (Note that the engine itself is quite noisy since I haven't dealt with the exhaust yet!)
  10. Mar 12, 2010 #9
    Just use a microphone touching the spring (just stick its face on the spring) and record the sounds at 44khz you can easily resolve milliseconds. The engine sounds will be background compared to that. So in the recording, you'll have some background followed by some large peaks (main) followed by repeating smaller decaying peaks(sound waves reverberating within the spring). You should be able to get the time from that.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook